An miRNA-Based Classification Model of Stages, Outcomes of ccRCC

By Kaitlyn D’Onofrio - Last Updated: June 8, 2021

MicroRNAs (miRNAs) are often used as biomarkers in certain cancers, but they are not widely used in clear cell renal cell carcinoma (ccRCC), the most common RCC subtype. Researchers evaluated a machine learning (ML) strategy to identify a novel miRNA signature predictive of stage and prognosis in patients with ccRCC in a recent study.

The researchers used best subset regression (BSR) analysis to determine the optimal prognostic model, and five ML algorithms were implemented to create stage classification models.

They found a four-miRNA signature that correlated closely with survival and was associated with high-risk patients, who, compared to the low-risk group, had unfavorable overall survival (hazard ratio, 4.523). This was confirmed in univariate and multivariate analyses. Using a combined ML algorithm, six miRNA signatures were found to be predictive of cancer staging. The Support Vector Machine algorithm had the best classification performance, including 0.923 accuracy, 0.927 sensitivity, and 0.919 specificity.

“A novel miRNA classification model using the identified prognostic and tumor stage–associated miRNA signature will be useful for risk and stage stratification for clinical practice, and the identified miRNA signature can provide promising insight to understand the progression mechanism of ccRCC,” the study authors concluded.